Notice: MemcachePool::get() [function.MemcachePool-get]: Server
xxx.xxx.xxx.xxx (tcp 0, udp 0) failed with: Failed to parse address
"xxx.xxx.xxx.xxx" (0) in /appdata/inventory/lib/db/
memcache_mysql.inc.php on line 16
I recently upgraded from 1.2.1 to 1.2.6 and this error now randomly
shows up. Upon doing an strace on the secondary server I see the
following errors...
epoll_wait(3, {{EPOLLIN, {u32=1732026752, u64=47710228750720}}}, 8196,
476) = 1
accept(23, {sa_family=AF_INET, sin_port=htons(51742),
sin_addr=inet_addr("xxx.xxx.xxx.xxx")}, [2218585775023325200]) = 30
fcntl(30, F_GETFL) = 0x2 (flags O_RDWR)
fcntl(30, F_SETFL, O_RDWR|O_NONBLOCK) = 0
write(19, "\0", 1) = 1
accept(23, 0x7fff437236f0, [2218585775023325312]) = -1 EAGAIN
(Resource temporarily unavailable)
I'm not sure whether this is a problem with the memcache php module or
at the memcached itself. This seems to happen when doing a flush.
After looking at the code for the memcache php module, it appears it
is looping through the list of servers and flushing the cache on each
one, but these errors pop up.
Versions
memcache php: 3.0.1
memcached: 1.2.6
CentOS 5U2
php 5.1.6
php.ini options for memcache (compiled as module)
; Memcache
memcache.redundancy = 1
memcache.session_redundancy = 2
Any insight would be helpful.
Thanks,
Rob
